Among the goals of this project are to describe adult age differences and changes in memory and learning performance and to investigate psychological processes underlying such age-related performance. This year an extensive longitudinal analysis of individual regression measures of change in vocabulary was carried out. It is generally accepted in gerontology that an individual's world knowledge does not decline with age. One of the ways such knowledge has been measured is with vocabulary tests. Longitudinal analysis of 12-year (3 points) and 18-year (4 points) changes in WAIS Vocabulary performance of the men in the Baltimore Longitudinal Study indicates small mean declines for the group who were in their sixties and seventies when measured initially. These declines together with the small means increases for the young groups produced a correlation of -.42 between age and change in vocabulary.